[benchmark] Fix quantile estimation type

[palimondo]

The correct quantile estimation type for printing all measurements in the summary report while quantile == num-samples - 1 is R-1, SAS-3. It's the inverse of empirical distribution function.

This enables printing all measured values in the one line of summary stats when executing Benchmark_O with --quantile parameter set to one less than --num-samples, without the need to resort to use of much heavier output in --verbose mode.

(The mistake was detected while porting the benchmark analysis to R-language. The previously chosen type R-3 behaves differently than I thought initially.)

References:

• https://en.wikipedia.org/wiki/Quantile#Estimating_quantiles_from_a_sample
• discussion in [benchmark] Report Quantiles from Benchmark_O and a TON of Gardening #19097 (comment)

[palimondo]

@eeckstein Please review 🙏
(… also please run a full test, since that's the one that executes the benchmarks integration tests)

[eeckstein]

@swift-ci test

[eeckstein]

@swift-ci benchmark

[eeckstein]

lgtm

[swift-ci]

Build comment file:

No performance and code size changes

How to read the data

The tables contain differences in performance which are larger than 8% and differences in code size which are larger than 1%.

If you see any unexpected regressions, you should consider fixing the regressions before you merge the PR.

Noise: Sometimes the performance results (not code size!) contain false alarms. Unexpected regressions which are marked with '(?)' are probably noise. If you see regressions which you cannot explain you can try to run the benchmarks again. If regressions still show up, please consult with the performance team (@eeckstein).

Hardware Overview

  Model Name: Mac Pro
  Model Identifier: MacPro6,1
  Processor Name: 12-Core Intel Xeon E5
  Processor Speed: 2.7 GHz
  Number of Processors: 1
  Total Number of Cores: 12
  L2 Cache (per Core): 256 KB
  L3 Cache: 30 MB
  Memory: 64 GB

---